2026-10-13 · 2026-Q4
Anthropic: Small Open-Weight Models Are Close to Self-Replicating Across Cheap IoT Devices
Outside the model: AI capabilityAgent tooling was not the attack surface, so no layer applies.
- Fixable
- Architectural
- Vendor response
- Unknown
- In the wild
- No
- Actor
- Not applicable
Target
Consumer IoT devices with unauthenticated ADB access, tested on emulated networks and 15 off-brand Android TV boxes
Impact
Anthropic's Frontier Red Team tested 11 open-weight models. GLM-5.2 propagated Llama3.1-8B to all 8 nodes of an emulated network in 4 of 5 trials, Qwen3.5-9B set up inference on 8 of 9 hardware types (45% success, 84% when earlier attempts were retained), and Gemma4-26B was split across 15 TV boxes at 0.39 tokens per second. Anthropic estimates models of 30B parameters or fewer could reach the full capability by the end of 2026 and found no working replicator in the wild. Research, not an incident
Root cause
Large fleets of cheap devices with exposed debug interfaces provide compute and spread paths for a model able to chain the steps itself
Sources
Catalog
Cite this: AgentProv incident record: Anthropic: Small Open-Weight Models Are Close to Self-Replicating Across Cheap IoT Devices (2026-10-13). https://agentprov.com/incidents/2026-10-13-anthropic-small-open-weight-models-are-close-to-self-replicating-across-cheap